Deep Learning in medical imaging (diagnosis, prediction) and clinical imaging outcomes prediction, clinical decision support, imaging utilization and appropriateness, cohort feature engineering with structured and unstructured EMR data for modeling applications
Our laboratory develops multidisciplinary engineering-based systems to study, target, and rewire cellular and multicellular circuits at scale, focusing on tumor immunology. We integrate genetic engineering tools with single cell, imaging, and spatial sequencing to perform massively parallel experiments, and use/develop machine learning and statistical inference to go from data to mechanisms and identify combinatorial effects within and across cells. Leveraging our multidisciplinary approach and capabilities we aim to identify immunomodulating interventions at an accelerated pace and lay the foundation for advances in disease diagnosis, treatment, and prevention.
Applications to Biology, in particular phylogenetic trees. Computational statistics, in particular, nonparametric computer intensive methods such as the bootstrap. Teaching using simulations and web-based tools. Image analysis. Immunology.
My research areas include statistical genetics, risk prediction modeling, cancer screening, and health policy modeling. I have been developing various statistical methods to analyze large-scale genetic data to understand the interplay between genes and the environment for various complex disease including cancer and neurological diseases. My recent methodological papers were published in high-profile statistical journals including the Journal of the American Statistical Association and Biometrics. In addition to statistical genetics, I have worked on several cancer screening and health policy modeling projects, by developing stochastic simulation models utilizing/merging various data sources including cancer registry data, epidemiologic case-control or cohort data, and nationally representative data such as NHANES (National Health and Nutrition Examination Survey). I have a wide range of methodological projects that BMI students may be interested in learning and working on. Currently, I am advising several medical students and Neurosurgery residents in conducting research, which includes: SEER-Medicare data-based surgery outcome analysis, mutation profiling for lung cancer using Stanford EMR database, and meta-analysis of
substance use disorders.